Fuzzy Relational Database Model and Managemant System for Imprecise Information

Abstract

This paper introduces a fuzzy relational database model (FRDB) and the management system for it. FRDB is built by extending the classical relational database model with the fuzzy membership degree of tuples in relations. The management system for FRDB with the querying language like SQL is built by using a classical open-source management system.

An Extended Type-2 Fuzzy Relational Database Model for Aggregate and Grouping Operations

“…Other FRDB models, such as [12], [13], [14], [15], [16] and [17], represented a fuzzy relation as a fuzzy set of tuples whose each attribute only took a single value and the membership degree was a real number in [0, 1] or a fuzzy number on [0, 1]. Fuzzy relational algebraic operations on these models were defined by extending directly classical relational algebraic operations based on fuzzy set operations.…”

“…Fuzzy relational algebraic operations on these models were defined by extending directly classical relational algebraic operations based on fuzzy set operations. A FRDB model is called a type-1 fuzzy relational database model (T-1FRDB), as in [12], [13] and [14], if the membership degree of a relation is a real number in [0, 1], and a type-2 fuzzy relational database model (T-2FRDB), as in [15], [16] and [17], if that is a fuzzy number on [0, 1]. T-2FRDB is better than T-1FRDB in representing imprecise information.…”

“…In [13], some forms of fuzzy aggregation queries were presented using SQL but there are no formal definitions for these forms. T-2FRDB models in [15], [16] and [17] were proposed to overcome shortcomings of T-1FRDB models in representing imprecise information, however fuzzy aggregate and grouping operations were not defined therein. Thus, the abilities of expressing and dealing with imprecise information of those T-2FRDB models were limited in the real world applications.…”